Esophageal cancer: pneumo-64-MDCT

Abdominal Imaging

ª Springer Science+Business Media, LLC 2009 Published online: 30 June 2009

Abdom Imaging (2010) 35:383–389 DOI: 10.1007/s00261-009-9554-3

Esophageal cancer: pneumo-64-MDCT Marina Ulla,1 Demetrio Cavadas,2 Ine´s Mun˜oz,1 Axel Beskow,2 Alberto Seehaus,1 Ricardo Garcı´a-Mo´naco1 1

Department of Radiology, Hospital Italiano de Buenos Aires, University of Buenos Aires, Potosı´ 4249 Buenos Aires 1181, Argentina 2 Department of Surgery, Hospital Italiano de Buenos Aires, University of Buenos Aires, Potosı´ 4249 Buenos Aires 1181, Argentina

Abstract

Introduction

Background: Pre-surgical characterization and staging of esophageal cancer with only one imaging method could be useful since gastric invasion determines the scope of resection and the kind of surgery to be performed. Our aim is to demonstrate the usefulness of Pneumo-64-MDCT (PnCT64) in the presurgical characterization of esophageal neoplasms in correlation with surgical findings. Materials and methods: A total of 50 patients with diagnosis of esophageal neoplasm were prospectively studied. A 14 French Foley catheter was used transorally in all patients. Air was instilled through the catheter to achieve esophageal distension. A 64-row MDCT scan was performed and the tumor was characterized according to scope, shape and anatomic location by using multiplanar 3D reconstructions and virtual endoscopy. Wall infiltration and presence of adenopathies were analyzed. Results: Adequate gastroesophageal distension was achieved in all patients. In 44/50 patients, wall thickening was observed, and in 34/50 regional adenopathies were found. In 29/50 patients the lesion was found in the lower third and in the gastroesophageal junction. The surgical correlation for wall infiltration was 85.7%. Conclusions: PnCT64 proved to be useful and safe for identification of esophageal wall thickening and presurgical characterization. Optimal distension allowed definition of both upper and lower borders of the tumors located in the gastroesophageal junction, of utmost importance to determine the surgical approach.

In recent decades, a constant increase in the prevalence of adenocarcinomas located in the lower third of the esophagus and the gastroesophageal (GE) junction has been observed. Therefore, the typical distribution of esophageal neoplasms has changed [1, 2]. This condition has typically been associated with high morbimortality rates, and a poor long-term survival rate, due to the late diagnosis of these lesions and the complex surgical approach required. The follow-up of patients with gastroesophageal reflux and Barrett’s esophagus has recently led to the detection of less advanced tumors, which in turn has improved long-term survival rates in patients with this condition [3]. The stage of the disease at the time of diagnosis and the possibility of performing radical resection (R0), i.e., with neither macro nor microscopic residual disease, are the most relevant factors for prognosis [3]. Upper endoscopy (UE) is the first study required to make a diagnosis and obtain the specimen for the lesion pathological diagnosis [4, 5]. However, this is not an accurate method to characterize the lesion and highgrade stenoses are difficult to be overcome by endoscopy. Barium studies are widely used and when performed by a skillful operator provide valuable information, but no data about tumor staging [4]. It is then necessary to use additional imaging studies for correct staging, which is essential to determine the therapeutic approach [4]. Since its emergence, computed tomography (CT) has been a key imaging examination for staging, since organ cross section clearly assesses the relationship of the esophageal tumor with other mediastinal structures and the presence of distant metastasis [5, 6]. Therefore, CT offers valuable information for therapeutic strategy and resectability assessment. However, conventional CT scans have some limitations for hollow organ assessment in the absence of

Key words: Esophagus—Cancer—64-MDCT— Characterization—Surgery

Correspondence to: Marina Ulla; email: marina.ulla@hospitalitaliano. org.ar

384

M. Ulla et al.: Esophageal cancer: pneumo-64-MDCT

Fig. 1. Adenocarcinoma at the GE junction (long arrow), with gastric fundus infiltration and left lung metastasis (short arrow). A Barium study. B–D Coronal MPR images. Irregular wall thickening of the esophageal wall with obliteration of the

adjacent fat tissue, and presence of adenomegaly adjacent to gastric fundus (arrowhead). E Virtual endoscopy. F Macroscopic specimen.

lumen distension, since the organ wall may be collapsed. Indeed, it is sometimes difficult to define wall thickening or to guarantee the presence of esophageal tumors with endoluminal growth. For these reasons, optimal esophageal distension could be very useful to overcome these limitations. Esophageal and gastric lumen distension may be performed with oral contrast agents, but it is often suboptimal due to contrast rapid transit. Moreover, oral contrast enhancement may generate confusing images, with the same density as the tumor [6, 7]. This potential drawback is crucial at the level of the GE junction, a typically difficult region to evaluate, where depiction of the tumor anatomic location conditions the surgical strategy. The recent introduction of multidetector computed tomography (MDCT) provided high quality isotropic images and volume reconstruction in both the coronal and sagittal planes, as well as virtual esophageal endoscopy. These technical advances contributed to a better anatomic depiction in few seconds scanning, thus

allowing easier tumor localization and subsequent surgical planning even for very small lesions [7–10]. In order to optimize tumor visualization in the esophageal wall and in the GE junction, a Foley catheter was used to achieve maximum lumen distension, which would better highlight the thickened areas in relation to the normal esophageal wall (Fig. 1). The aim of this article was to show the usefulness of this PnCT64 technique, in the presurgical assessment and characterization of esophageal and GE junction neoplasms.

Materials and methods Between April 2007 and July 2008, 50 patients (32 males, 18 females), mean age 64 years (range 37–84 years), with an endoscopically confirmed diagnosis of esophageal neoplasm were studied. Patients were admitted by a nurse to the radiology department with 8 h fasting and a peripheral venous line placed. Once at the MDCT suite,

M. Ulla et al.: Esophageal cancer: pneumo-64-MDCT

385

Fig. 2. Adenocarcinoma at the GE junction (long arrow) with cardia infiltration and local and regional adenomegalies (short arrow). A Barium study. B–C Axial images. D–E Coronal and

sagittal MPR evidencing tumor scope and wall thickening with polypoid and ulcerated areas. Transparency window 3D reconstruction. G UE. H Virtual endoscopy. I Open surgical specimen.

a Foley 14 French catheter was placed transorally under the cricopharyngeal muscle. Then MDCT was performed before and 40 and 80 s after intravenous injection. Nonionic iodine contrast (Omnipaque 300; Nycomed Amersham, Princeton, NJ) at a dose of 1 mL/kg was infused using an automatic injection pump at a flow of 3.5 mL/s. No oral contrast was used. Anterior and lateral scout views were obtained to program the different tomographic phases. In each of the tomographic acquisitions, air was manually instilled through the Foley catheter in a continuous and sustained fashion to achieve good esophageal distension. Each acquisition lasted an average of 8 s. An Aquilion 64-row MDCT (Toshiba Inc,

Tokyo, Japan) was used for examination with 0.5 mm slices, 0.25 mm table feed, 50 mAs, 120 kV, 0.75 s rotation time and 0.875 pitch. Once acquired, images were sent to a Vitrea 2 working station (Vital Image Inc.) for evaluation. Simple and curved multiplanar reconstruction (MPR) in different planes was performed with lung and soft tissue windows. Volume rendering and transparency window reconstructions generating images similar to those of a barium study were performed. Moreover, virtual esophageal endoscopy was done (Fig. 2). These reconstructions were used to assess all wall thickening over 3 mm that was considered pathological.

M. Ulla et al.: Esophageal cancer: pneumo-64-MDCT

386

Fig. 3. Adenocarcinoma at the GE junction (long arrow) with gastric fundus involvement and lower curvature associated with sliding hiatus hernia (short arrow). A Sagittal MPR

depicting tumor scope and parietal thickening. B–C Volume rendering and transparency window 3D reconstructions.

Table 1. Distribution of esophageal wall thickening

Surgery was performed in 35/50 patients: 3 had early stage tumors (not shown by PnCT64), and 32 had advanced lesions (Table 2). The type of surgery performed was as follows: 31 total esophagectomies, 3 total gastroesophageal resections, and 1 limited gastroesophageal resection. Pathology findings revealed 24/35 adenocarcinomas and 11/35 squamous carcinomas. Comparison of PnCT64 findings and histopathological findings of patients undergoing resection showed a surgical correlation for wall infiltration in 85.7% (30/35). In 5/35 operated patients, PnCT64 examination did not evidence wall infiltration: 3 were T1 tumors, and the other 2 were T2 tumors. The surgical correlation for the detection of adenopathies reached 75%. Of 15/50 patients who did not undergo surgery, 1 had early stage tumors (not shown in PnTC64), and 14 had advanced stage tumors (8 received palliative care and stenting, while the other 6 were referred to chemoradiotherapy) (Table 2).

No evidence of thickening Upper third thickening Middle third thickening Lower third thickening Total patients

6 3 12 29 50

Table 2.

Underwent surgery Underwent no surgery Total

No. of patients

Advanced tumors

Early tumors

35 15 50

32 14*, 46

3 1 4

 

* 8 patients treated with stent   6 patients who underwent chemo-radiotherapy

Lesion shape, size, and anatomic location were characterized (Fig. 3). Obliteration of the fat planes and presence of adenomegalies were reported (Fig. 2). MDCT images were correlated with surgical findings.

Results Both esophageal and gastric distension was optimal in all the studies (n = 50). Patients developed no complications or adverse events, except for one patient who complained of mild chest pain, but required no treatment. PnCT64 showed esophageal wall thickening (over 3 mm) in 44/50 patients (84%). The neoplasm was detected in the lower third and GE junction in 29/44, in the middle third in 12/44, and in the upper third in 3/44 (Table 1). Regional adenopathies were observed in 34/50 (68%) patients.

Discussion Early diagnosis and accurate staging of esophageal cancer are both essential for therapeutic strategy planning [3]. UE with biopsy is usually the study of choice to confirm the diagnosis; however, it provides no information about cancer staging. Barium studies are useful to characterize stenosis and assess the size in the organ longitudinal axis. This is a key factor to define gastric fundus involvement in neoplasms involving the GE junction, since the stomach is the organ

M. Ulla et al.: Esophageal cancer: pneumo-64-MDCT

387

Fig. 4. Squamous carcinoma at the GE junction (long arrow) associated with gastric polyp (short arrow). A–B Coronal and sagittal MPR reconstructions evidencing both focused wall

thickening and polyp. C–D Transparency window 3D reconstruction. E–F Virtual endoscopy.

usually used as the first choice for reconstruction after esophagectomy [3, 4]. Other imaging studies such as CT, magnetic resonance imaging (MRI), positron emission tomography (PET), and endoscopic ultrasound have proved useful for tumor assessment and staging in patients with malignant esophageal stenosis [4, 7, 11, 12]. Thoraco-abdominal CT is frequently used as a first step for local and distal staging of esophageal cancer. However, the esophageal walls in conventional CT scans are usually collapsed. Therefore, endoluminal and wall thickening assessment is virtually impossible [6, 7]. On the contrary, adequate esophageal lumen distension evidences and characterizes the intraluminal and intramural pathology, as well as the shape, anatomic location and size of intraluminal masses, or esophageal wall thickening [6–8, 10, 13, 14]. It also enables a better visualization of the extraluminal planes: fat layers between the tumor and the adjacent organs, mediastinal perforation, and tracheo-bronchial tree involvement [5]. The advantages of MDCT over conventional CT consist in the acquisition of extremely high quality images in all planes, as well as 3D reconstructions and virtual endoscopic views [15, 16], as seen in this series (Fig. 4). Other authors have previously evaluated the role of 3D reconstructions and virtual endoscopy of MDCT in both presurgical assessment and esophageal cancer assessment, and commented on their advantages over con-

ventional CT [1, 6]. In these reports esophageal distension was achieved with granular sparkling and/or oral contrast agents [6, 17]. To our knowledge, including a Medline search, there are no previous reports about PnCT64 using the technique herein described. Although the use of a transoral Foley catheter may be rather uncomfortable, the distension obtained was optimal in all cases, contrary to what was previously reported with granular sparkling or oral contrast agents, where the required esophageal distension was not always achieved [6, 17]. Moreover, this technique was useful to adequately distend not only the esophagus but also the stomach, thus defining the lower limit of the lesion in tumors invading the GE junction. This information is of great value for surgeons, since gastric wall involvement may radically change the surgical approach [3, 18, 19]. Some authors believe that the best surgery for these lesions located in the GE junction is esophagectomy and total gastrectomy with total colonic reconstruction, as it was performed in three patients in our series [3, 18, 19]. However, this practice is questionable, since no better survival rate has been proven with this extended resection. Indeed, in case a good and safe surgical margin can be obtained with a different type of surgery, it would not be wise to perform esophageal resection or colon reconstruction [19, 20]. Stomach distension led to an adequate definition of both the upper and lower borders

388

M. Ulla et al.: Esophageal cancer: pneumo-64-MDCT

Fig. 5. Adenocarcinoma located in the middle third of the esophagus (arrow). A–B Coronal and sagittal MPR reconstructions evidencing thickening localization and scope. C

MiniMip reconstruction. D–E: 3D reconstructions. F Open macroscopic specimen.

of the lesion in tumors located in the GE junction, which in turn was useful to plan the surgical approach. Once the presence of distant metastasis has been ruled out, the detailed information provided by PnCT64 is an excellent tool to determine the presence of a locally advanced lesion. This is defined by the involvement of both the esophagus and the stomach in GE junction tumors, or by the presence of mediastinal and celiac adenopathies [19]. In this setting, neoadjuvancy is the best therapeutic strategy. The association of presurgical chemotherapy and radiotherapy has led to disease understaging in many patients, thus favoring radical surgery and improving the long term survival rate [19]. In esophageal tumors considered irresectable due to the presence of distant metastasis or local invasion, or in high-risk patients, palliation therapy with self expandable stenting is the method of choice. In this series, the definition of both the upper and lower limits of the

tumor in the longitudinal axis provided by PnCT64 allowed to determine the stent graft length and the need for a valved stent, and no further barium studies were required. According to the results obtained in this series, PnCT64 as a sole imaging method was useful for staging and therapeutic strategy. A high correlation was observed of PnCT64 findings with histopathological findings in patients with wall thickening over 3 mm, who underwent resection. But it has shown no benefit in early lesions. Compared with previous reports describing a different technique for esophageal distension [6], this series evidenced a better correlation for wall infiltration, which could be due to the optimal distension obtained with an esophageal catheter and a 64MDCT. Endoscopic ultrasound has proved to be a very useful tool and a highly sensitive method for local T and N staging. It also plays a key role in the detection of early stage tumors for the assessment of lesion invasion and

389

M. Ulla et al.: Esophageal cancer: pneumo-64-MDCT

the presence of regional adenopathies. Data provided by endoscopic ultrasound may lead to either endoscopic or surgical treatment, including regional lymphadenectomy vs. traditional resection with extensive lymphadenectomy [19–22]. However, endoscopic ultrasound has some drawbacks, such as limitations to detect distant lymph nodes involvement or metastatic disease, or to overcome a high grade stenosis. In this series, endoscopic ultrasound was not systematically used for advanced lesion staging, since it would not have changed the therapeutic approach [5, 20, 21]. Other imaging modalities, such as MR and PET scans, are not used systematically to study esophageal cancer, except in particular circumstances or in search of post-treatment relapse [11, 12], and have been scarcely performed in this series. In our study, PnCT64 provided precise anatomical location, detailed characterization and accurate staging in esophageal cancer as a result of panoramic multiorgan assessment, endoscopic views and both multiplanar and 3D reconstructions (Fig. 5). Imaging evaluation in the coronal plane proved to be of great value given the easy localization of subcarinal adenopathies and complete assessment of the upper and lower borders of the tumor, which in turn helped design a therapeutic strategy (Fig. 1). Virtual endoscopy images might be a useful choice for patients with incomplete endoscopies due to high grade esophageal stenosis, or for patients with a poor clinical condition. Some study limitations are the lack of comparison of PnCT64 with other imaging modalities, including endoscopic ultrasound and MR. In analyzing this series there might have been a bias, because the radiologist knew that the patient had an already confirmed esophageal neoplasm.

Conclusion PnCT64 proved to be a useful, safe and accurate technique to identify esophageal wall thickening and to stage esophageal cancer in this series of patients. The additional stomach distension led to an adequate definition of both the upper and lower borders of the lesion in tumors located in the GE junction, which in turn was helpful to design the surgical approach. Acknowledgment. To Cora Smith and Valeria Melia for English assistance in writing this article.

References 1. Devesa S, Blot W, Fraumeni J, et al. (1998) Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 83:2049–2053 2. Pera M (2003) Trends in incidence and prevalence of specialized intestinal metaplasia, Barrett’s esophagus and adenocarcinoma of the gastroesophageal junction. World J Surg 27:999–1008 3. Siewert JR, Feith M, Stein H (2005) Biologic and clinical variations of adenocarcinoma at the esophagogastric junction: relevance of a topographic anatomic subclassification. J Surg Oncol 90:139–146 4. Kumbasar B (2002) Carcinoma of esophagus: radiologic diagnosis and staging. Eur J Radiol 42:170–180 5. Onbas O, Eroglu A, Kantarci M, et al. (2006) Preoperative staging of esophageal carcinoma with multidetector CT and virtual endoscopy. Eur J Radiol 57:90–95 6. Mazzeo S, Caramella D, Gennai A, et al. (2004) Multidetector CT and virtual endoscopy in the evaluation of the esophagus. Abdom Imaging 29:2–8 7. Kim AY, Kim HJ, Ha HK (2005) Gastric cancer by multidetector row CT: preoperative staging. Abdom Imaging 30(4):465–472 8. Prokop M (2003) Multislice CT: technical principles and future trends. Eur Radiol 5:3–13 9. Rydberg J, Liang Y, Teague SD (2004) Fundamentals of multichannel CT. Semin Musculoskelet Radiol 8:137–146 10. Kohl G (2005) The evolution and state-of-the-art principles of multislice computed tomography. Proc Am Thorac Soc 2(6):470– 476, 499–500 11. Lightdale CJ (2000) Positron emission tomography: another useful test for staging esophageal cancer. J Clin Oncol 15(18):3199–3201 12. Flanagan FL, Dehdashti F, Siegel BA, et al. (1997) Staging of esophageal cancer with 18F-fluorodeoxyglucose positron emission tomography. AJR Am J Roentgenol 168:417–424 13. Carrascosa P, Capun˜ay C, Martin Lopez, et al. (2008) Esophageal stenosis; three dimensional multidetector CT and virtual endoscopy. Abdom Imaging 34(1):19–25 14. Thompson WM (1997) Esophageal carcinoma. Abdom Imaging 22:138–142 15. Mahesh M (2002) Search for isotropic resolution in CT from conventional through multiple-row detector. Radiographics 22(4):949–962 16. Hara AK, Johnson CD, MacCarty RL, et al. (2001) CT colonography: single versus multi-detector row imaging. Radiology 219(2):461–465 17. Grazheani H, Panebianco V, Iafrate F, et al. (2006) 3D CT protocol in the assessment of esophageal neoplastic lesions: can it improve TNM staging? Eur Radiol 16:414–421 18. Millikan KW, Silverstein J, Hart V, et al. (1995) A 15-year review of esophagectomy for carcinoma of the esophagus and cardia. Arch Surg 130:617–624 19. William A, Killinger Jr, Thomas W, et al. (1996) Stage II esophageal carcinoma: the significance of T and N. J Thorac Cardiovasc Surg 111(5):935–940 20. Boyce GA, Rice TW, Sivak MV (1991) Esophageal ultrasound and the preoperative staging of carcinoma of the esophagus. J Thorac Cardiovasc Surg 101:536 21. Stein H, Feith M, Bruecher B, Naehrig J, et al. (2005) Early esophageal cancer: pattern of lymphatic spread and prognostic factors for long-term survival after surgical resection. Ann Surg 242:566–573 22. Stein H, Hutter J, Feith M, et al. (2007) Limited surgical resection and jejunal interposition for early adenocarcinoma of the distal esophagus. Semin Thorac Cardiovasc Surg 19:72–78